To download the eBook with 6 Killer Innovations in Data Analytics click here.
Killer Innovation 1 – You Don’t Need to be a Data Scientist to Predict the Future
Predictive Analytics allows you to predict a future outcome based on past data and knowledge. This is realised using a combination of Data Science and Computer Science called Machine Learning. Essentially an algorithm uses historical data to find relationships and patterns in the data, and build a predictive model. New data is then fed through this model and it predicts outcomes.
Azure Machine Learning is a cloud service that lets you easily design, test, operationalise and manage predictive analytics solutions. Build your model, train it on sample data, verify it, and then let it produce results for you on real data.
You don’t need to be a data scientist to work with and deploy this – you can think of Azure Machine Learning as “Applied” Machine Learning which is simple for developers or data professionals to use. The only thing you need is a web browser and an internet connection – everything is provided as part of the service.
Once you’ve done all your experimentation and testing and have optimised your model, you can deploy your model into production as a web service in minutes all from within the Machine Learning Service. Anything can call this web service, you provide the input, it runs it through the model and produces the output which you can then use in any way you want.
An example of anomaly detection using Machine Learning can be found in Credit Card Fraud Detection: Analyse real time credit card transactions for customers and identify anomalous, potentially fraudulent transactions and then take action accordingly.